利用一个spark receiver 接收多个flume agent

String host = args[0];

  int port = Integer.parseInt(args[1]);

  String host1 = args[2];

  int port1 = Integer.parseInt(args[3]);

  InetSocketAddress address1 = new InetSocketAddress(host,port);

  InetSocketAddress address2 = new InetSocketAddress(host1,port1);

  InetSocketAddress[] InetSocketAddressArray = {address1,address2};

JavaStreamingContext jssc = new JavaStreamingContext(new SparkConf().setAppName("JavaFlumeEventHandle_1"), Durations.seconds(2));

 JavaReceiverInputDStream<SparkFlumeEvent> flumeStream = FlumeUtils.createPollingStream(jssc, InetSocketAddressArray, StorageLevel.MEMORY_AND_DISK_SER_2());

2个spark receiver接收多个flume agent

        String host = args[0];

        int port = Integer.parseInt(args[1]);

        String host1 = args[2];

        int port1 = Integer.parseInt(args[3]);

        InetSocketAddress address1 = new InetSocketAddress(host,port);

        InetSocketAddress address2 = new InetSocketAddress(host1,port1);

        InetSocketAddress[] InetSocketAddressArray = {address1,address2};

      JavaStreamingContext jssc = new JavaStreamingContext(new SparkConf().setAppName("JavaFlumeEventHandle_1"), Durations.seconds(2));

//       JavaReceiverInputDStream<SparkFlumeEvent> flumeStream = FlumeUtils.createPollingStream(jssc, InetSocketAddressArray, StorageLevel.MEMORY_AND_DISK_SER_2());

        JavaReceiverInputDStream<SparkFlumeEvent> flumeStream = FlumeUtils.createPollingStream(jssc,host,port);

        JavaReceiverInputDStream<SparkFlumeEvent> flumeStream1 = FlumeUtils.createPollingStream(jssc, host1, port1);

        JavaDStream<SparkFlumeEvent> union = flumeStream.union(flumeStream1);

性能参照： 

http://blog.sina.com.cn/s/blog_13122bdbc0102vay7.html